RLWM ACT-R RMSE model fitting and outcome analysis for split-half data

Results

Model fits

#>   model     name freq
#> 1   LTM half1_N3   70
#> 2   LTM half1_N6   63
#> 3   LTM half2_N3   64
#> 4   LTM half2_N6   59
#> 5    RL half1_N3   13
#> 6    RL half1_N6   20
#> 7    RL half2_N3   19
#> 8    RL half2_N6   24

#> # A tibble: 8 × 4
#> # Groups:   half, condition [4]
#>   half  condition model `length(subjects)`
#>   <chr> <chr>     <chr>              <int>
#> 1 half1 set3      LTM                   70
#> 2 half1 set3      RL                    13
#> 3 half1 set6      LTM                   63
#> 4 half1 set6      RL                    20
#> 5 half2 set3      LTM                   64
#> 6 half2 set3      RL                    19
#> 7 half2 set6      LTM                   59
#> 8 half2 set6      RL                    24

Dynamics

Does the best fit model change from 1st half to second half and between the two set-sizes?

Are the parameter values largely different for each half and set-size?

Individual plots